Modeling Multivariate Time Series in Economics: From Auto-Regressions to Recurrent Neural Networks

A new paper by Sergiy Verstyuk:

Abstract: The modeling of multivariate time series in an agnostic manner, without assumptions about underlying theoretical structure is traditionally conducted using Vector Auto-Regressions. They are well suited for linear and state-independent evolution. A more general methodology of Multivariate Recurrent Neural Networks allows to capture non-linear and state-dependent dynamics. This paper takes a range of small- to large-scale Long Short-Term Memory MRNNs and pits them against VARs in an application to US data on GDP growth, inflation, commodity prices, Fed Funds rate and bank reserves. Even in a small-sample regime, MRNN outperforms VAR in forecasting: its out-of-sample predictions are about 20% more accurate. MRNN also fares better in interpretability by means of impulse response functions: for instance, a temporary shock to the Fed Funds rate variable generates system dynamics that are more plausible according to conventional

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3297736

 

Big Data 2018 Videos

On August 23-24, 2018 the CMSA hosted our fourth annual Conference on Big Data. The Conference featured many speakers from the Harvard community as well as scholars from across the globe, with talks focusing on computer science, statistics, math and physics, and economics. Videos of the talks are contained in the youtube playlist below.

You can visit the event page for more information.

 

Streetchange Won a Webby Award!

Streetchange won a Webby award for Best Use of Machine Learning on the Web! Congratulations to the team.

Streetchange is a new way of measuring changes in the physical appearances of neighborhoods using a computer vision algorithm. The researchers calculated Streetchange by algorithmically comparing Google Street View images of the same location captured in different years.

Read more about the project here.

Read the paper here.

Check out the award page here.

Videos from Big Data 2017

The Big Data Conference featured many speakers from the Harvard community as well as scholars from across the globe, with talks focusing on computer science, statistics, math and physics, and economics. This was the third conference on Big Data the Center hosted as part of our annual events, and was co-organized by Richard Freeman, Scott Kominers, Jun Liu, Horng-Tzer Yau and Shing-Tung Yau.

This youtube playlist contains all of the videos from the conference. For detailed abstracts and slides, you can find the schedule here.